Transformation of 2,4,6-trinitrotoluene (TNT) by ... - Springer Link

13 downloads 0 Views 270KB Size Report
trinitrotoluene (TNT) and other nitroorganic compounds were investigated for their ... One isolate, Raoultella terrigena strain HB, removed TNT at concentrations ...
 Springer 2006

Biodegradation (2007) 18:27–35 DOI 10.1007/s10532-005-9033-7

Transformation of 2,4,6-trinitrotoluene (TNT) by Raoultella terrigena H. Claus1,*, T. Bausinger2, I. Lehmler1, N. Perret3, G. Fels3, U. Dehner2, J. Preuß2 & H. Ko¨nig1 1

Institute of Microbiology and Wine Research, Johannes Gutenberg-University Mainz, J. J. Becherweg 15, D-55099 Mainz, Germany; 2Institute of Geography, Johannes Gutenberg-University Mainz, J. J. Becherweg 21, D-55099 Mainz, Germany; 3Faculty of Sciences, Department of Chemistry, University of Paderborn, Warburger Str. 100, D-33098 Paderborn, Germany (*author for correspondence: e-mail: hclaus@ uni-mainz.de) Accepted 18 November 2005

Key words: azoxy-dimers, bioremediation, Raoultella terrigena, TNT

Abstract Manufacture of nitroorganic explosives generates toxic wastes leading to contamination of soils and waters, especially groundwater. For that reason bacteria living in environments highly contaminated with 2,4,6trinitrotoluene (TNT) and other nitroorganic compounds were investigated for their capacity for TNT degradation. One isolate, Raoultella terrigena strain HB, removed TNT at concentrations between 10 and 100 mg l)1 completely from culture supernatants under optimum aerobic conditions within several hours. Only low concentrations of nutrient supplements were needed for the cometabolic transformation process. Radioactivity measurements with ring-labelled 14C–TNT detected about 10–20% of the initial radioactivity in the culture supernatant and the residual 80–90% as water-insoluble organic compounds in the cellular pellet. HPLC analysis identified aminodinitrotoluenes (2-ADNT, 4-ADNT) and diaminonitrotoluenes (2,4DANT) as the metabolites which remained soluble in the culture medium and azoxy-dimers as the main products in the cell extracts. Hence, the new isolate could be useful for the removal of TNT from contaminated waters. Abbreviations: ADNT – aminodinitrotoluene; DANT – diaminonitrotoluene; TN-4,4¢-azo – 2,2¢,6,6¢-tetranitro-4,4¢-azotoluene; TN-2,2¢-azoxy – 4,4¢,6,6¢-tetranitro-2,2¢-azoxytoluene; TN-4,4¢-azoxy – 2,2¢,6,6¢tetranitro-4,4¢-azoxytoluene; TNT – 2,4,6-trinitrotoluene

Introduction 2,4,6-Trinitrotoluene (TNT) is a nitroaromatic explosive that has been released into soil and water ecosystems mainly due to its massive use during the two World Wars. As a result, many sites used for TNT production have become seriously contaminated with the explosive and related compounds (Fuller et al. 2004; Lewis et al. 2004). Typical explosive-contaminated sites may contain up to 10,000 mg kg)1 TNT in soils and up to 100 mg l)1 in water. TNT and its metabolites exhibit a high toxic and mutagenic potential on prokaryotes and eukaryotes (Spanggord et al.

1995; Honeycutt et al. 1996; Lachance et al. 1999). It has been estimated that nearly 3,200 sites in Germany require environmental cleanup (Preuß & Haas 1987; Preuß & Eitelberg 1999). Biological based remediation strategies are promising economical and ecological alternatives for existing physical/chemical technologies. The extensive research on biodegradation of TNT by bacteria and fungi has been summarized in several recent reviews (Fritsche et al. 2000; Hawari et al. 2000; Lenke et al. 2000; Rosser et al. 2001; van Aken & Agathos 2001; Heiss & Knackmus 2002; Zhao et al. 2004). The presence of three electron-withdrawing nitro groups in TNT

28 introduces steric constraints and confers a high electron deficiency to the aromatic ring. The molecule is thus regarded as resistant to oxidative microbial degradation, although some bacteria can use the nitro-groups of TNT as a nitrogen source. Only low mineralization rates have been reported for bacteria in contrast to several ligninolytic fungi. Instead of oxidation, many bacteria catalyze the reduction of one or two nitro groups of TNT to monoaminodinitrotoluenes (ADNT) and diaminonitrotoluenes (DANT) under both aerobic and anaerobic conditions. The electron transfer is mediated by oxygen-insensitive cytoplasmatic nitroreductases (Pak et al. 2000; Kim & Song 2005). Reactive nitroso- and hydroxylamino intermediates can further react to condensated azoxy-dimers and acetyl derivates of TNT. Under strictly anaerobic conditions ADNT is further reduced to 2,4,6-triaminotoluene (TAT) which is highly reactive, and can polymerize or irreversibly bind to soil (Thiele et al. 2002). The reductive reactions are the basis of several treatment processes for the bioremediation of TNT contaminated soils (Lenke et al. 2000; Fuller et al. 2004; Lewis et al. 2004). No comparable biological approach exists for contaminated aquatic environments. In this study we describe a new isolate Raoultella terrigena strain HB originating from a contaminated former explosive production site, as a vehicle to remove TNT from contaminated waters.

Materials and methods Isolation of TNT-degrading microorganisms Water and soil samples were collected from abandoned TNT production sites in Germany (Hallschlag/Rhineland-Palatinate, Moschwig/SaxonyAnhalt). Microorganisms were enriched in 0.1Standard I nutrient broth (Merck, Darmstadt, Germany, No. 107882) supplemented with TNT (10 mg l)1). From these cultures single colonies were obtained on Standard I nutrient agar (Merck, Darmstadt, Germany, No. 107881) containing TNT (10 mg l)1) and further characterized. TNT used for degradation experiments originated from a solid sample collected at a former explosives production site in Germany; it was recrystallized from absolute ethanol. In the

resulting product no impurites could be detected by GC and HPLC. The identity was confirmed by determination of the melting point, retention times, UV-spectrum and EI-mass spectrum against a reference sample of TNT (Dynamit Nobel, Troisdorf, Germany). Identification of the isolates Isolates of different morphological colony appearance were further differentiated by macrorestriction and plasmid analysis as described by Claus et al. (1995). Identification was done by sequencing of the 16S rDNA genes amplified by polymerase chain reaction (PCR). Primers used were Eubak 338F (ACTCCTACGGGAGGCAG) and C 1992R (CCACGGGCGGTGTGTAC). The PCR amplificates were sequenced by Genterprise Inc. (Mainz, Germany). Analytical methods Reversed-phase high-performance liquid chromatography (HPLC) was used to separate TNT and other nitroorganic compounds. Before HPLC analyses all samples were centrifuged (16,000g, 5 min) and filtrated (membrane pore size: 0.2 lm). The following HPLC-DAD equipment was used (all parts Gynkotek, Germany): pump M 480, on-line degasser, autosampler Gina 50, diode array detector UVD 340 S, column oven STH. The separation was performed on a Nucleodur column 100–3 C18ec 2504 mm (Macherey-Nagel, Du¨ren, Germany) using a multistep methanol/water gradient: 1. step: 0 fi 10 min, 40 fi 50% methanol; 2. step: 10 fi 35 min, isocratic 50% methanol; 3. step: 35 fi 80 min, 50 fi 85% methanol at a flow rate of 0.4 ml min)1 and a column temperature of 20 C. The injection volume was 50 ll, the detection wavelength was set to 230 nm. For the separation of the three azo- and azoxy-derivatives a Nucleosil column 120–3 C18ec 2504 mm (Macherey-Nagel, Du¨ren, Germany) and an acetonitrile/water gradient was used (0 fi 60 min, 40 fi 85% acetonitrile). The flow rate and the column temperature were identical to the first method. The following analytical standards were used: 2-amino-4,6-dinitrotoluene (2-ADNT), 4amino-2,6-dinitrotoluene (4-ADNT), 2,4-diamino6-nitrotoluene (2,4-DANT), (LGC Promochem, Wesel, Germany); 3,5-dinitroaniline (3,5-DNA),

29 2-nitroaniline (2-NA), 1,3,5-trinitrobenzene (TNB) (Sigma-Aldrich, Taufkirchen, Germany); 1,3dinitrobenzene (1,3-DNB), 2,4-dinitrotoluene (2,4-DNT) (Merck, Darmstadt, Germany); hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX), 2,4,6-trinitrotoluene (TNT) (Dynamit Nobel, Troisdorf, Germany); 2,2¢,6,6¢-tetranitro-4,4¢-azotoluene (TN-4,4¢-azo) (AccuStandard, New Haven, USA); 4,4¢,6,6¢-tetranitro-2,2¢-azoxytoluene (TN-2,2¢-azoxy), 2,2¢,6,6¢-tetranitro-4,4¢-azoxytoluene (TN-4,4¢-azoxy) (synthesis according to Sitzmann 1974). Samples of 14C-activity were analyzed by HPLCUV with a flow-through scintillation counter (Radio-HPLC) on a Merck Hitachi model LaChrom 7000 series equipped with an UV-detector set at 254 nm and a Synergi 4u MAX-RP 80A 2504.6 mm (Phenomenex) column. The radioactivity was measured with an EG&G Berthold model LB 507 B apparatus using a measuring cell YG 150U 4. A binary gradient elution program was used consisting of water (A) and methanol (B): 1. step: 0 fi 50 min, 25 fi 100% B; 2. step: 50 fi 60 min, isocratic 100% B; 3. step: 60 fi 70 min, 100 fi 25% B; 4. step: 70 fi 80 min, isocratic 25% B. The flow was kept constant at 0.7 ml min)1. For rapid estimations of microbial TNT degradation a spectrophotometric method was adapted from Oh et al. (2000). Nitrite in the culture fluids was measured spectrophotometrically as described by Kim et al. (2002). TNT degradation experiments The mineral salt medium of Kalafut et al. (1998) was used to study the TNT degrading capabilities of the bacterial isolates. TNT was added to the mineral salt medium in concentrations of 10, 50 and 100 mg l)1 before autoclaving (15 min, 121 C; traces of TNB were sometimes identified after this treatment). For experiments with solid media, mineral salt media were prepared with agarose (12 g l)1). The influence of nutrient supplements was tested at increasing concentrations (0–3.0%) of Standard I nutrient broth or glucose (0–1.0%) in the minimal salt medium. In order to simulate more natural conditions, the mineral salt media were also prepared with contaminated water and soil extracts from the original site of the microbial isolates. The soil

extracts were obtained by shaking a 10% (w/v) aqueous soil suspension for 2 h (200 rpm) before autoclaving (20 min, 121 C). Water insoluble materials were removed by centrifugation (10,000g, 30 min) and the supernatant was passed through a paper filter (Schleicher and Schuell, Dassel, Germany, No. 589). The media were prepared either in volumes of 50 ml and 100 ml in Erlenmeyer flasks or 3 ml in 16 ml screw cap tubes. Cells used for inocula were precultured in Standard I nutrient broth for 16 h at 30 C on a shaker. The cells were either used directly or washed twice in mineral salt media before inoculation into mineral salt media. The cultures were incubated up to 7 days at 30 C on a rotary shaker (200 rpm). At regular intervals, aliquots were taken from the cultures to determine bacterial growth and TNT elimination. Growth of the cultures was followed by counting the colony forming units (cfu) on Standard I nutrient agar and by measuring the optical density at 600 nm. TNT and its metabolites in the culture supernatants were determined after centrifugation of the aliquots at 16,000g for 5 min. At the end of the experiments, cells and waterinsoluble materials were separated by centrifugation at 40,000g for 30 min. The resulting pellet was washed twice with phosphate buffered saline (pH 7.4). The pellet was extracted with acetonitrile for 16 h at 30 C and centrifuged as above. The resulting fractions (supernatant, washings, acetonitrile extract) were analysed by HPLC. A separate set of experiments was performed with resting cells: 1 ml samples of a preculture grown in Standard I nutrient broth were centrifuged (16,000g, 5 min) and the cell pellets resuspended in the same volume of minimal salt medium containing TNT (100 mg l)1). After incubation at 30 C the assays were centrifuged (16,000g, 5 min) at the indicated intervals and the supernatants analysed by HPLC. Cells inactivated by autoclaving served as controls. Mass balance of

14

C-TNT

[Ring U)14C]TNT (2.2 mCi mmol)1) was synthesised as described by Kro¨ger and Fels (2000). Radioactivity was determined by liquid scintillation counting (1414 Wallac WinSpectral Liquid Scintillation Counter, Turku, Finland, and

30 TriCarb Liquid Scintillation Analyzer, Model 2500TR, Packard Canberra) with Rotiszint eco plus (Roth, Karlsruhe, Germany, No. 0016.2) as cocktail. Minimal salt media were prepared with nonradioactive TNT (100 mg l)1) and up to 2106 dpm ml)1 [ring U-14C]TNT. The nutrient media (100 ml) were inoculated with 500 ll of a preculture in Standard I nutrient broth and incubated for up to 7 days at 30 C on a rotary shaker (200 rpm). Cells and water-insoluble sediments were separated by centrifugation at 40,000g for 30 min and washed twice with phosphate buffered saline (pH 7.4). Radioactivity in the different fractions (culture supernatant, washings, cells, water-insoluble sediment) was determined after mixing 100 ll sample with 4 ml of Rotiszint eco plus. C14-radioactivity was also measured in the acetonitrile extract of the sediment (obtained as above). Unextractable 14C remaining in the pellet was determined by combusting to 14CO2 in a biological oxidizer (Zinsser OX 500).

Results Isolation and characterization of TNT transforming bacteria Microorganisms were isolated from water and soil samples of two former ammunition production plants in Germany. Bacteria from these contaminated environments were isolated on Standard I nutrient agar in the presence of TNT (10 mg l)1). One isolate which transformed TNT most efficiently was identified by the sequence of its 16S rDNA (700 nucleotides sequenced) as Raoultella terrigena. Strain HB (DSMZ No. 16101) consists of Gram-negative, oxidase-negative, catalasepositive, non-motile rods. It grows well at temperatures between 4 C and 40 C with an optimum at 30 C (data not shown). Growth at low temperatures is a hallmark of the new described genus Raoultella, which is facultative anaerobic, having both a respiratory and a fermentative type of metabolism (Drancourt et al. 2001).

brownish pigments within and around the growth zone (Figure 1). Growth in liquid media was determined by colony counts and optical density. After a lag time of 24 h, the colony forming units (cfu) in the minimal salt medium with TNT increased rapidly and reached the same level as in the medium without TNT (Figure 2). The significant increase of the optical density of cells grown in the presence of TNT is not a result of higher cell densities, but obviously attributed to altered spectroscopic properties of the cells and the culture media by accumulation of coloured TNT metabolites, similar to those observed on solid agar media. Determination of the optical density is thus not an appropriate parameter for estimating bacterial growth in dependence of TNT. Cometabolic transformation R. terrigena strain HB did not grow with TNT as the only source of carbon and energy and no TNT transformation was observed without an additional nutrient source. However, microbial growth and transformation of TNT occurred even at low nutrient concentrations in the minimal salt media (Table 1). At concentrations of ‡0.05% Standard I nutrient broth or glucose in the minimal salt media, TNT was completely removed from the culture broth within 7 days of incubation. When the inocula were taken directly from a preculture grown in Standard I nutrient broth without further cell washings, the nutrients supplied by this complex medium were sufficient to

Growth in the presence of TNT On minimal salt agar supplemented with TNT (100 mg l)1), R. terrigena strain HB produced

Figure 1. Colonies of R. terrigena strain HB on solid minimal salt media containing TNT (100 mg l)1); Dark (brownish) pigments accumulated within and around the growth zones.

31 1.00E+10

2 1.8 1.6

1.00E+09

cfu/ml

1.2 1

1.00E+08

0.8

OD (600 nm)

1.4

0.6 1.00E+07 0.4 0.2 1.00E+06

0 0

1

2

3

4

5

6

7

8

days

Figure 2. Growth of R. terrigena strain HB in liquid minimal salt media without (D) and with (n) 100 mg TNT l)1; solid line (colony counts); dotted line (optical density).

promote bacterial TNT transformation (Table 2). Under the same conditions TNT was removed from original surface waters or soil extracts contaminated with a mixture of nitroorganic compounds (Table 2). With pregrown (resting) cells, the cometabolic transformation of TNT occurred within a few hours of incubation (Figure 3). No reaction was Table 1. Influence of nutrient concentrations on bacterial TNT elimination* Nutrient

Concentration in minimal salt medium (%)

TNT (mg l)1) in cultures of R. terrigena HB

Glucose

0.000 0.003 0.006 0.012 0.050 0.000 0.004 0.008 0.016 0.032

89 88 40 37